Gliomas are the most common malignant human brain tumors, accounting for the majority of deaths from primary brain neoplasms. Even when treated with surgery, radiotherapy, and chemotherapy, patients with grade IV glioblastoma (GB) have a life expectancy of only 14 months. A major challenge to treating GB is its highly invasive nature, as infiltrating cancer cells are ?protected? from exposure to systemically administered chemo- and immunotherapies by the blood brain barrier (BBB). To overcome this physical this barrier to drug delivery, we non-invasively open the BBB via the activation of intravascular microbubbles (MBs) with MRI-guided focused ultrasound (FUS). In this proposal, we will utilize this non-invasive approach to deliver anti-CD47 immunotherapy (mCD47) to gliomas. CD47, a surface molecule expressed by cancer cells that sends a so-called ?don?t eat me? signal to phagocytic cells, has garnered considerable attention recently because functionally blocking CD47 elicits control of many neoplasms, including primary brain tumors. Our proposal has 2 specific aims. In Aim 1, we will determine, through the use of 89Zr-ImmunoPET imaging, how the timing and sequencing of FUS application with respect to drug administration affects mCD47 delivery to GL261 gliomas. BBB opening with FUS and MBs is extremely complex, with several factors potentially influencing mAb delivery (e.g. time-varying modulation of tumor interstitial pressure, active vs. passive transport, and altered efflux transporter expression levels). Thus, there is strong rationale to definitively establish how the timing and sequencing of FUS with respect to drug administration affects the delivery of mCD47. Next, Aim 2 will investigate the experimental therapeutic efficacy of mCD47 against gliomas when delivered using the most effective timing and sequencing regimen identified in Aim 1. We will evaluate the ability of FUS to improve mCD47 efficacy in controlling GL261 tumor growth, blocking infiltration, and improving survival. Reproducibility will be tested using the CT-2A glioma model, which is also both highly infiltrative and responsive to mCD47. By mid-2019, dozens of patients worldwide had been treated with FUS and MBs to elicit BBB opening, thus a clear precedent has been set for translation of this technology. UVa houses a low-frequency clinical MR image-guided FUS system (Insightec Exablate Neuro) that is specialized for this application, and we are preparing for our first clinical trial for drug delivery across the BBB in GB patients using FUS and MBs. Thus, if we are able to demonstrate therapeutic efficacy of mCD47 in combination with FUS and MBs, we are exceptionally well-positioned to rapidly translate these findings to the clinic.